The goal of this proposal is to use new knowledge on endocytosis to optimize the cellular effects of anti-cancer immunotoxins. Once targeted to the surface of tumor cells, immunotoxins must be internalized and undergo an intracellular activation process that includes the reduction of their interchain disulfide bond. Almost nothing is known about reductive cleavage of endocytosed disulfides. The key importance of this step is demonstrated by the fact that immunotoxins lose effectiveness when their disulfide is replaced by a thioether bond. Biochemical measurements of this reductive cleavage have eluded many efforts, in part because of the lack of suitable probes that could be used as models for disulfide linked toxins. We have developed such probes by linking an efficiently endocytosed and proteolytically undegradable polymer (poly(D-lysine)), through disulfide linkage either to an anticancer drug (methotrexate) to obtain a cytocidal conjugate (MTX-SS-PDL) or to 125I-tyramine to obtain a labeled analogue (125I-tyn-SS-PDL) used in a simple cleavage assay. Disulfide cleavage in CHO cells pulse labeled with 125I-tyn-SS-PDL begins without lag and continues for several hours. Initial cleavage at the cell surface (first 15-30 minutes) is followed by intracellular cleavage which, according to our preliminary data on cell fractionation, occurs in the Golgi apparatus. Our experimental approach will be: 1) to clarify the functional significance of the surface cleavage; 2) to confirm the reductive function of the Golgi apparatus; 3) to study the mechanisms of reductive cleavage, using among others: blockers of surface sulfhydryls, antibodies against enzymes suspected to catalyze the disulfide reduction, inhibitors of golgi function and intracellular transport, an in vitro assay of Golgi function, mutants of CHO cells defective in their ability to cleave endocytosed disulfide bonds and mutants hypersensitive to ricin; factors known to increase ricin cytotoxicity. cystamine, which we found to be a strong enhancer of ricin cytotoxicity will be studied with particular attention. An improved biochemical assay and a new functional assay of cell-mediated ricin cleavage are being developed and will be used to correlate cleavage and cytotoxicity, to pinpoint the steps of intracellular processing (uptake, routing, activation, translocation) that insure optimal ricin toxicity, and to determine which of these steps are subject to experimental modulation. Two immunotoxins in which ricin A is disulfide- linked to monoclonal antibodies against cell surface proteins of CHO cells will be prepared. Since their surface targets are internalized by different forms of endocytosis (receptor-mediated versus non-receptor- mediated) the comparison will clarify which of the two insures more efficient delivery of activated ricin to its cellular target. Correlating the data obtained with the two poly(lysine) conjugates, ricin and two ricin immunotoxins, is expected to provide a rational basis for an optimal clinical use of imunotoxins.